job-posting-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted job posting data from external sources, creating a potential surface for indirect prompt injection.
- Ingestion points: Job description text provided by users or fetched from external URLs as instructed in
SKILL.md. - Boundary markers: The skill enforces a strict analysis process using deterministic Python scripts and requires the output to follow the template in
references/report-format.md. - Capability inventory: The skill executes a local Python script (
scripts/analyze_job_posting.py) to calculate hourly rates and decompose salary components. - Sanitization: The
scripts/_common.pyutility provides rigorous input validation, ensuring data conforms to expected JSON types and checking for the presence of PII (emails, phone numbers, and MyNumber IDs) using regex patterns.
Audit Metadata